The Datastore Definition section of a script defines the source and target datastores including Cursor objects.
- Source datastores are most commonly url:// style references to SQDaemon processes running on remote systems and the Capture/Publisher agents serviced by the daemon. They may also be local or remote databases or files accessible on the system's local file system.
- Target datastores may also be url:// style references that identify a supported datastore API or local or remote databases or files accessible on the local file system.
- Cursor objects are local or remote databases most commonly used to validate foreign key references or retrieve and/or update subsets of data related to the Source or Target Datastores.
The physical sequence of the items that define Source and Target datastores is critical in that they must be defined in the script prior to performing any processing against them. In addition the DESCRIPTION(s) of the objects in the source and target datastores must precede the DATASTORE(s) and CREATE RDBMS CURSOR(s) Commands that reference them.
- DESCRIPTION – identifies the layout structure (i.e. copybooks, XML DTDs, relational DDL, etc.) of a record within a datastore.
- DATASTORE – identifies the source and target databases/files/queues, etc that will be processed by the Apply Engine.
- CREATE RDBMS CURSOR - defines a relational database cursor that can perform any legal SQL operation. The Apply Engine can access only one Relational datastore at a time as defined by the RDBMS command. Cursors can validate Keys or retrieve rows using full or partial keys. They can provide access to data not available in the Source datastore and perform one-to-one, one-to-many and many-to-one datastore operations. A Cursor can provide query and update access to a third datastore if neither the Source or Target datastore is relational.